5af13607-7741-41ac-b110-c94c8a3ee0f0

pending completion

Build # 5af13607-7741-41ac-b110-c94c8a3ee0f0

Build Type

Pull #441

circleci

Committed by

hectorcorrea

Commit Message

Better error handling

Pull Request Pull Request #441: Indexing to a new collection

Run Details

99 of 99 new or added lines in 3 files covered. (100.0%)

1816 of 2220 relevant lines covered (81.8%)

95.45 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

91.0

/app/models/dataset_citation.rb

# frozen_string_literal: true

# Handles citations for datasets
# rubocop:disable Metrics/ParameterLists
# rubocop:disable Metrics/ClassLength
# rubocop:disable Style/NumericPredicate
# rubocop:disable Style/IfUnlessModifier
class DatasetCitation
  NEWLINE_INDENTED = "\r\n\t\t\t\t\t\t\t\t"

  # @param authors [<String>] Array of authors.
  # @param years [<String>] Array of years (expected 1 or 2 values).
  # @param title [String>] Title of the dataset
  # @param type [String] Type of the dataset (e.g. "Data set" or "Unpublished raw data")
  # @publisher [String] Publisher of the dataset
  # @doi [String] DOI URL
  def initialize(authors, years, title, type, publisher, doi)
    @authors = authors || []
    @years = years || []
    @title = title
    @type = type
    @publisher = publisher
    @doi = doi
  end

  def to_s(style)
    if style == "BibTeX"
      bibtex
    else
      apa
    end
  end

  # Returns a string with APA-ish citation for the dataset
  # Reference: https://libguides.usc.edu/APA7th/datasets#s-lg-box-22855503
  def apa
    apa_author = ''
    case @authors.count
    when 0
      # do nothing
    when 1
      apa_author += @authors.first
    when 2
      apa_author += @authors.join(' & ')
    else
      apa_author += @authors[0..-2].join(', ') + ', & ' + @authors[-1]
    end

    apa_year = ''
    case @years.count
    when 0
      # do nothing
    when 1
      apa_year += "(#{@years.first})"
    else
      apa_year += "(#{@years.join('-')})"
    end

    apa_title = DatasetCitation.custom_strip(@title)
    apa_title += " [#{@type}]" if @type.present?
    apa_title = append_dot(apa_title)

    apa_publisher = append_dot(@publisher)
    apa_doi = @doi

    tokens = [append_dot(apa_author), append_dot(apa_year), apa_title, apa_publisher, apa_doi].reject(&:blank?)
    tokens.join(' ')
  rescue => ex
    Rails.logger.error "Error generating APA citation for (#{@title}): #{ex.message}"
    nil
  end

  # Returns a string with BibTeX citation for the dataset.
  #
  # There is no set standard for datasets and therefore the format we produce
  # was put together from a variety of sources including mimicking what Zotero
  # does and looking at examples from Zenodo (e.g. https://zenodo.org/record/6062882/export/hx#.Yiejad9OnUL)
  #
  # Notice that we use the @electronic{...} identifier instead of @dataset{...} since
  # Zotero does not recognize the later.
  def bibtex
    tokens = []
    if @authors.count > 0
      tokens << bibtex_field_author('author', @authors)
    end

    if @title.present?
      tokens << bibtex_field('title', @title, '{{', '}}')
    end

    if @publisher.present?
      tokens << bibtex_field('publisher', @publisher, '{{', '}}')
    end

    if @years.count > 0
      tokens << bibtex_field('year', @years.first)
    end

    if @doi.present?
      tokens << bibtex_field('url', @doi, '{', '}')
    end

    text = "@electronic{#{bibtex_id},\r\n"
    text += tokens.map { |token| "\t#{token}" }.join(",\r\n") + "\r\n"
    text += "}"
    text
  rescue => ex
    Rails.logger.error "Error generating BibTex citation for (#{@title}): #{ex.message}"
    nil
  end

  # Return a string with the ContextObjects in Spans (COinS) information
  # https://en.wikipedia.org/wiki/COinS
  def coins
    tokens = []
    tokens << "url_ver=Z39.88-2004"
    tokens << "ctx_ver=Z39.88-2004"
    tokens << "rft.type=webpage"
    tokens << "rft_val_fmt=#{CGI.escape('info:ofi/fmt:kev:mtx:dc')}"

    if @title.present?
      tokens << "rft.title=#{CGI.escape(@title)}"
    end

    @authors.each do |author|
      tokens << "rft.au=#{CGI.escape(author)}"
    end

    if @years.count > 0
      tokens << "rft.date=#{CGI.escape(@years.first.to_s)}"
    end

    if @publisher.present?
      tokens << "rft.publisher=#{CGI.escape(@publisher)}"
    end

    if @doi.present?
      tokens << "rft.identifier=#{CGI.escape(@doi)}"
    end

    "<span class=\"Z3988\" title=\"#{tokens.join('&amp;')}\"></span>"
  rescue => ex
    Rails.logger.error "Error generating COinS citation for (#{@title}): #{ex.message}"
    nil
  end

  # Returns an ID value for a BibTex citation
  def bibtex_id
    author_id = 'unknown'
    if @authors.count > 0
      author_id = @authors.first.downcase.tr(' ', '_').gsub(/[^a-z0-9_]/, '')
    end
    year_id = @years.first&.to_s || 'unknown'
    "#{author_id}_#{year_id}"
  end

  # Breaks a string into lines of at most max_length.
  # Returns an array with the lines.
  def bibtex_lines(string, max_length = 40)
    string = string.to_s # handles non-string values gracefully
    lines = []
    until string.nil?
      # TODO: it would be nice it we break on spaces rather than in the middle of a word.
      lines << string[0..max_length - 1]
      string = string[max_length..-1]
    end
    lines
  end

  # Creates a line to represent a BibTex field.
  # Breaks long lines into smaller lines.
  # Examples:
  #
  #     field_name = {{ short value }}
  #     field_name = {{ very very very
  #                 very very very very
  #                 long value }}
  #
  def bibtex_field(name, value, open_tag = '', close_tag = '')
    value_trim = bibtex_lines(value).join(NEWLINE_INDENTED)
    name.ljust(12) + '= ' + open_tag + value_trim + close_tag
  end

  # Creates a line to represent multiple authors in a BibTex field
  # https://en.wikibooks.org/wiki/LaTeX/Bibliography_Management#Authors
  #
  # Example:
  #
  #     author = { author1 and
  #              author2 and
  #              author3 }
  #
  def bibtex_field_author(name, authors, open_tag = '{', close_tag = '}')
    value_trim = authors.join(" and #{NEWLINE_INDENTED}")
    name.ljust(12) + '= ' + open_tag + value_trim + close_tag
  end

  # Appends a dot to a string if it does not end with one.
  def append_dot(value)
    return nil if value.nil?
    return '' if value.empty?
    DatasetCitation.custom_strip(value) + '.'
  end

  # Strip a few specific characters that tend to mess up citations (e.g. trailing periods, commas, et cetera)
  def self.custom_strip(value)
    return nil if value.nil?
    return '' if value.empty?
    while true
      last_char = value[-1]
      break if last_char.nil? || !last_char.in?('. ,')
      value = value.chomp(last_char)
    end
    value
  end
end
# rubocop:enable Metrics/ParameterLists
# rubocop:enable Metrics/ClassLength
# rubocop:enable Style/NumericPredicate
# rubocop:enable Style/IfUnlessModifier

1	# frozen_string_literal: true
2
3	# Handles citations for datasets
4	# rubocop:disable Metrics/ParameterLists
5	# rubocop:disable Metrics/ClassLength
6	# rubocop:disable Style/NumericPredicate
7	# rubocop:disable Style/IfUnlessModifier
8	class DatasetCitation	1✔
9	NEWLINE_INDENTED = "\r\n\t\t\t\t\t\t\t\t"	1✔
10
11	# @param authors [<String>] Array of authors.
12	# @param years [<String>] Array of years (expected 1 or 2 values).
13	# @param title [String>] Title of the dataset
14	# @param type [String] Type of the dataset (e.g. "Data set" or "Unpublished raw data")
15	# @publisher [String] Publisher of the dataset
16	# @doi [String] DOI URL
17	def initialize(authors, years, title, type, publisher, doi)	1✔
18	@authors = authors \|\| []	8✔
19	@years = years \|\| []	8✔
20	@title = title	8✔
21	@type = type	8✔
22	@publisher = publisher	8✔
23	@doi = doi	8✔
24	end
25
26	def to_s(style)	1✔
27	if style == "BibTeX"	×
28	bibtex	×
29	else
30	apa	×
31	end
32	end
33
34	# Returns a string with APA-ish citation for the dataset
35	# Reference: https://libguides.usc.edu/APA7th/datasets#s-lg-box-22855503
36	def apa	1✔
37	apa_author = ''	5✔
38	case @authors.count	5✔
39	when 0
40	# do nothing
41	when 1
42	apa_author += @authors.first	3✔
43	when 2
44	apa_author += @authors.join(' & ')	1✔
45	else
46	apa_author += @authors[0..-2].join(', ') + ', & ' + @authors[-1]	1✔
47	end
48
49	apa_year = ''	5✔
50	case @years.count	5✔
51	when 0
52	# do nothing
53	when 1
54	apa_year += "(#{@years.first})"	4✔
55	else
56	apa_year += "(#{@years.join('-')})"	1✔
57	end
58
59	apa_title = DatasetCitation.custom_strip(@title)	5✔
60	apa_title += " [#{@type}]" if @type.present?	5✔
61	apa_title = append_dot(apa_title)	5✔
62
63	apa_publisher = append_dot(@publisher)	5✔
64	apa_doi = @doi	5✔
65
66	tokens = [append_dot(apa_author), append_dot(apa_year), apa_title, apa_publisher, apa_doi].reject(&:blank?)	5✔
67	tokens.join(' ')	5✔
68	rescue => ex
69	Rails.logger.error "Error generating APA citation for (#{@title}): #{ex.message}"	×
70	nil	×
71	end
72
73	# Returns a string with BibTeX citation for the dataset.
74	#
75	# There is no set standard for datasets and therefore the format we produce
76	# was put together from a variety of sources including mimicking what Zotero
77	# does and looking at examples from Zenodo (e.g. https://zenodo.org/record/6062882/export/hx#.Yiejad9OnUL)
78	#
79	# Notice that we use the @electronic{...} identifier instead of @dataset{...} since
80	# Zotero does not recognize the later.
81	def bibtex	1✔
82	tokens = []	1✔
83	if @authors.count > 0	1✔
84	tokens << bibtex_field_author('author', @authors)	1✔
85	end
86
87	if @title.present?	1✔
88	tokens << bibtex_field('title', @title, '{{', '}}')	1✔
89	end
90
91	if @publisher.present?	1✔
92	tokens << bibtex_field('publisher', @publisher, '{{', '}}')	1✔
93	end
94
95	if @years.count > 0	1✔
96	tokens << bibtex_field('year', @years.first)	1✔
97	end
98
99	if @doi.present?	1✔
100	tokens << bibtex_field('url', @doi, '{', '}')	1✔
101	end
102
103	text = "@electronic{#{bibtex_id},\r\n"	1✔
104	text += tokens.map { \|token\| "\t#{token}" }.join(",\r\n") + "\r\n"	6✔
105	text += "}"	1✔
106	text	1✔
107	rescue => ex
108	Rails.logger.error "Error generating BibTex citation for (#{@title}): #{ex.message}"	×
109	nil	×
110	end
111
112	# Return a string with the ContextObjects in Spans (COinS) information
113	# https://en.wikipedia.org/wiki/COinS
114	def coins	1✔
115	tokens = []	1✔
116	tokens << "url_ver=Z39.88-2004"	1✔
117	tokens << "ctx_ver=Z39.88-2004"	1✔
118	tokens << "rft.type=webpage"	1✔
119	tokens << "rft_val_fmt=#{CGI.escape('info:ofi/fmt:kev:mtx:dc')}"	1✔
120
121	if @title.present?	1✔
122	tokens << "rft.title=#{CGI.escape(@title)}"	1✔
123	end
124
125	@authors.each do \|author\|	1✔
126	tokens << "rft.au=#{CGI.escape(author)}"	1✔
127	end
128
129	if @years.count > 0	1✔
130	tokens << "rft.date=#{CGI.escape(@years.first.to_s)}"	1✔
131	end
132
133	if @publisher.present?	1✔
134	tokens << "rft.publisher=#{CGI.escape(@publisher)}"	1✔
135	end
136
137	if @doi.present?	1✔
138	tokens << "rft.identifier=#{CGI.escape(@doi)}"	1✔
139	end
140
141	"<span class=\"Z3988\" title=\"#{tokens.join('&')}\"></span>"	1✔
142	rescue => ex
143	Rails.logger.error "Error generating COinS citation for (#{@title}): #{ex.message}"	×
144	nil	×
145	end
146
147	# Returns an ID value for a BibTex citation
148	def bibtex_id	1✔
149	author_id = 'unknown'	1✔
150	if @authors.count > 0	1✔
151	author_id = @authors.first.downcase.tr(' ', '_').gsub(/[^a-z0-9_]/, '')	1✔
152	end
153	year_id = @years.first&.to_s \|\| 'unknown'	1✔
154	"#{author_id}_#{year_id}"	1✔
155	end
156
157	# Breaks a string into lines of at most max_length.
158	# Returns an array with the lines.
159	def bibtex_lines(string, max_length = 40)	1✔
160	string = string.to_s # handles non-string values gracefully	8✔
161	lines = []	8✔
162	until string.nil?	8✔
163	# TODO: it would be nice it we break on spaces rather than in the middle of a word.
164	lines << string[0..max_length - 1]	9✔
165	string = string[max_length..-1]	9✔
166	end
167	lines	8✔
168	end
169
170	# Creates a line to represent a BibTex field.
171	# Breaks long lines into smaller lines.
172	# Examples:
173	#
174	# field_name = {{ short value }}
175	# field_name = {{ very very very
176	# very very very very
177	# long value }}
178	#
179	def bibtex_field(name, value, open_tag = '', close_tag = '')	1✔
180	value_trim = bibtex_lines(value).join(NEWLINE_INDENTED)	4✔
181	name.ljust(12) + '= ' + open_tag + value_trim + close_tag	4✔
182	end
183
184	# Creates a line to represent multiple authors in a BibTex field
185	# https://en.wikibooks.org/wiki/LaTeX/Bibliography_Management#Authors
186	#
187	# Example:
188	#
189	# author = { author1 and
190	# author2 and
191	# author3 }
192	#
193	def bibtex_field_author(name, authors, open_tag = '{', close_tag = '}')	1✔
194	value_trim = authors.join(" and #{NEWLINE_INDENTED}")	1✔
195	name.ljust(12) + '= ' + open_tag + value_trim + close_tag	1✔
196	end
197
198	# Appends a dot to a string if it does not end with one.
199	def append_dot(value)	1✔
200	return nil if value.nil?	20✔
201	return '' if value.empty?	20✔
202	DatasetCitation.custom_strip(value) + '.'	20✔
203	end
204
205	# Strip a few specific characters that tend to mess up citations (e.g. trailing periods, commas, et cetera)
206	def self.custom_strip(value)	1✔
207	return nil if value.nil?	33✔
208	return '' if value.empty?	32✔
209	while true	31✔
210	last_char = value[-1]	53✔
211	break if last_char.nil? \|\| !last_char.in?('. ,')	53✔
212	value = value.chomp(last_char)	22✔
213	end
214	value	31✔
215	end
216	end
217	# rubocop:enable Metrics/ParameterLists
218	# rubocop:enable Metrics/ClassLength
219	# rubocop:enable Style/NumericPredicate
220	# rubocop:enable Style/IfUnlessModifier

pulibrary / pdc_discovery / 5af13607-7741-41ac-b110-c94c8a3ee0f0

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous